INDEX

Explanations

information related to specific options or choices

oai_token-act-pair · gpt-3.5-turbo

responses indicating approval or disapproval regarding various categories

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uyomi

-0.69

isconsin

-0.63

 destro

-0.63

 centrif

-0.63

viz

-0.60

 therap

-0.59

 advance

-0.58

":[

-0.58

 arrang

-0.57

 Mellon

-0.55

POSITIVE LOGITS

Unknown

0.71

asma

0.68

unknown

0.67

ixel

0.67

lam

0.64

tek

0.64

:\

0.64

CHAT

0.63

Height

0.62

 Requires

0.61

Activations Density 0.286%

information related to specific options or choices

responses indicating approval or disapproval regarding various categories

No Comments

No Known Activations